Skip Navigation Archive: U.S. Department of Health and Human Services U.S. Department of Health and Human Services
Archive: Agency for Healthcare Research Quality www.ahrq.gov
Archival print banner

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: https://info.ahrq.gov. Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to www.ahrq.gov for current information.

National Healthcare Disparities Report, 2013

Chapter 4. Patient Safety

Importance

Mortality
Number of Americans who die in hospitals each year from medical errors (1999 est.) 44,000-98,000 (Kohn, et al., 2000)i
Prevalence
Rate of harms associated with hospital stays (2000-2007) 25.1 per 100 admissions (Landrigan, et al., 2010)
Number of preventable adverse events among adults (excluding obstetrics) per year in U.S. hospitals (2004 est.) 3,023,000 (Jha, et al., 2009)
All-payer 30-day readmission rate 14.4% of admissions (HHS, 2012)
Cost
Cost of preventable adverse events for adults (nonobstetric) in U.S. hospitals (2013 est.) $22 billion (Jha, et al., 2009 adjusted)ii
Total cost per error in U.S. hospitals (2013 est.) $15,000 (Shreve, et al., 2010 adjusted)

Measures

The Institute of Medicine (IOM) defines patient safety as "freedom from accidental injury due to medical care or medical errors" (Kohn, et al., 2000). In 1999, the IOM published To Err Is Human: Building a Safer Health System, which called for a national effort to reduce medical errors and improve patient safety.

Measuring and tracking patient safety incidents is a necessary step to improving quality of care. Measuring patient safety is complicated by difficulties in ensuring the systematic reporting of patient safety incidents in ongoing, protected, consistent, and informative ways. For example, health care providers may fear that if they participate in the analysis of patient safety incidents, the findings may be used against them in court or harm their professional reputations.

Aggregating data that are defined differently across facilities or State lines is fraught with scientific difficulties, such as:

  • Counting the relevant subpopulation for calculating rates of error.
  • Having sufficient numbers to identify prevalent risks and hazards in the delivery of patient care.
  • Having the detail to identify underlying causes of these events and practices that are most effective in mitigating risks.

A combination of administrative data, medical record abstraction, voluntary adverse event reporting, and patient surveys is needed to understand what is and is not improving.

Despite these challenges, progress has been made in raising awareness, passing legislation, developing reporting systems, establishing national data collection standards, and conducting research:

  • The Joint Commission's sentinel event program, established in 1996, signaled to hospitals that accreditation depends on their timely review of unexpected death or serious injury and mitigation of such risks. It is believed that hospitals underreport sentinel events to the Joint Commission.
  • President George W. Bush signed the Patient Safety and Quality Improvement Act of 2005 to spur the development of voluntary, provider-driven initiatives to improve the quality, safety, and outcomes of patient care.
  • As of 2009, 27 States developed voluntary or mandatory reporting systems for, at a minimum, serious reportable events. These are adverse events that should never happen to a patient (NASHP, 2013).
  • The Agency for Healthcare Research and Quality (AHRQ) has certified 76 Patient Safety Organizations (PSOs) as of November 18, 2013. PSOs work to develop learning communities in patient safety and collect patient safety event reports that are legally protected from legal disclosure when reported to a PSO. This new program shifts from a culture of blame to a learning collaboration among providers and patient safety experts.
  • AHRQ has also developed a set of "common formats" for health care facilities and professionals to report patient safety events, near misses, and unsafe conditions to PSOs in a consistent way for aggregation and learning.
  • The National Quality Strategyiii is the result of collaborations among private and public organizations and aims to increase access to high-quality, affordable health care for all Americans, by spurring health care providers to reduce rates of care-related injury to zero when possible and redesign systems that reliably provide high-quality health care.
  • The Partnership for Patients, created by the U.S. Department of Health and Human Services, has set goals to make care safer.
  • More than 100 studies, using myriad data sources, have addressed patient safety progress made since the IOM report (Raetzman, et al., 2012).

Based on these efforts, this 2013 National Healthcare Disparities Report (NHDR) presents a number of patient safety measures organized around the major health care settings that must measure, understand, and improve health care in order for Americans to be cared for in a safer health care environment:

Hospital setting:

  • Hospital-acquired conditions overall.
  • Postoperative sepsis.
  • Catheter-associated urinary tract infections (UTIs).
  • Central line-associated bloodstream infections (CLABSIs).
  • Mechanical adverse events associated with central venous catheters.
  • Obstetric trauma.

Nursing home setting:

  • Pressure ulcers, use of restraints, and UTIs.

Home health setting:

  • Improvement in surgical site wound healing.
  • Ability to take medications orally.

Ambulatory care setting:

  • Ambulatory visits due to adverse effects of medical care.
  • Receipt of inappropriate prescription medications.
  • Hospital readmissions.

Infrastructure:

  • Patient safety event reporting in Pennsylvania. 
  • Patient safety culture.

Findings

Hospital Setting

Outcome: Overall Hospital-Acquired Condition Rate

Patient safety events that occur in the hospital setting are referred to as hospital-acquired conditions (HACs). Hospitals are a common setting for patient safety events in part because of the clinically compromised state of many patients admitted to the hospital and because of the high volume of care transactions and interventions that take place during a hospital stay.

Hospitalized Blacks are at higher risk than Whites of experiencing nosocomial infections and adverse drug events due to anticoagulants and hypoglycemic agents (Metersky, et al., 2011). Higher rates of these patient safety events also occur in hospitals serving high percentages of Black patients.

A key goal of the Federal Government's Partnership for Patients program is to make hospital care safer by reducing the rate of preventable HACs. To track progress on this goal, a method for estimating HACs was developed based on 28 different measures from three national data sources: Medicare Patient Safety Monitoring Systemiv (MPSMS) implemented by AHRQ and the Centers for Medicare & Medicaid Services (CMS), Patient Safety Indicators (PSIs) implemented by AHRQ, and the National Healthcare Safety Network (NHSN) implemented by the Centers for Disease Control and Prevention (CDC). The rate developed is intended to reflect the entire population over age 17.

Although the 28 measures have been combined, the overall HAC rate is not an all-inclusive HAC rate. Some important types of adverse events are not included in the rate due to a lack of data. For example, the rate does not include adverse drug events due to allergies or use of narcotics, venous thromboembolic events in nonsurgical patients, or most infections that are hospital acquired but do not produce symptoms until after hospital discharge. In addition, retained surgical items and wrong-site surgeries are not included.

Figure 4.1. Distribution of hospital-acquired conditions, based on national rates per 1,000 adult hospital discharges, by race/ethnicity, 2010

Select [D] Text Description below for details.

Select [D] Text Description below for details.

[D] Select for Text Description.

Source: Agency for Healthcare Research and Quality, Medicare Patient Safety Monitoring System, 2010; Centers for Disease Control and Prevention, National Healthcare Safety Network, 2009-2011; Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project, State Inpatient Databases and AHRQ Quality Indicators, version 4.1, 2011.
Note: Data are for patients age 18 years and over. Estimates are rounded to the nearest tenth.

  • The overall HAC rate includes nine specific HACs, as well as an "other" category that covers an additional 14 specific HACsv (Figure 4.1). In 2010, the overall HAC rate for Blacks was 148 per 1,000 hospital discharges compared with a rate for Whites of 143 per 1,000 hospital discharges.
  • For both Blacks and Whites, the HAC categories with the highest rates in 2010 were adverse drug events, pressure ulcers, and all other HACs.
  • Among Blacks, the 2010 rate of adverse drug events is roughly double that of pressure ulcers (63.4 and 30.2 per 1,000 hospital discharges, respectively). Among Whites, the rates for adverse drug events and for pressure ulcers are more similar (47.4 and 43.3 per 1,000 hospital discharges, respectively).

Also, in the National Healthcare Quality Report (NHQR):

  • In 2011, the national overall HAC rate was 142 per 1,000 hospital discharges. By comparison, the rate was 145 per 1,000 hospital discharges in 2010.

Healthcare-Associated Infections

Infections acquired during hospital care (nosocomial infections) are one of the most serious patient safety concerns. They are the most common complication of hospital care (Gastmeier, 2004). Approximately 1 out of every 20 hospitalized patients will contract an HAI (CDC, 2010). Annual costs for adult inpatients that are attributable to the five HAIs with the highest impact on the health care system (CLABSIs, surgical site infections, catheter-associated UTIs, ventilator-associated pneumonia, and Clostridium difficile infections ) are estimated at $9.8 billion (Zimlichman, et al., 2013).

A specific medical error cannot be identified in most cases of HAIs. However, better application of evidence-based preventive measures can reduce HAI rates within an institution.

Outcome: Postoperative Sepsis

Sepsis is a potentially life-threatening bloodstream infection that can be acquired in various settings. One study of sepsis occurring in community settings as well as hospital settings found that both higher rates of infection and higher risk of acute organ dysfunction contributed to higher rates of sepsis seen among Blacks compared with Whites (Mayr, et al., 2010). Sepsis can occur after surgery, and another recent study showed that postoperative sepsis occurred in 5% of emergency surgery patients and 2% of elective surgery patients (Moore, et al., 2010). One way that sepsis rates can be reduced is by giving patients appropriate prophylactic antibiotics starting 1 hour prior to surgical incision.

Figure 4.2. Postoperative sepsis per 1,000 adult discharges with an elective operating room procedure, by race/ethnicity and insurance, 2008-2010

Select [D] Text Description below for details.

 

Select [D] Text Description below for details.

[D] Select for Text Description.

Key: API = Asian or Pacific Islander.
Source: Agency for Healthcare Research and Quality (AHRQ), Healthcare Cost and Utilization Project, Nationwide Inpatient Sample and AHRQ Quality Indicators, modified version 4.1, State Inpatient Databases disparities analysis files, 2008-2010.
Denominator: All elective hospital surgical discharges, age 18 and over, with length of stay of 4 or more days, excluding patients admitted for infection, patients with cancer or immunocompromised states, patients with obstetric conditions, and admissions specifically for sepsis.
Note: For this measure, lower rates are better. White, Black, and API are non-Hispanic. Hispanic includes all races. Rates are adjusted by age, gender, age-gender interactions, comorbidities, and diagnosis-related group clusters.

  • In all years, Whites had a lower risk-adjusted rate of postoperative sepsis than Blacks and Hispanics. The rate for Whites was also lower than for Asians and Pacific Islanders (APIs) in 2009 and 2010 (Figure 4.2).
  • In 2008, only Medicaid patients had a higher risk-adjusted rate of postoperative sepsis than private insurance patients. Surgery patients with Medicare or Medicaid had higher rates than surgery patients with private insurance in 2009 and 2010.
  • The 2008 top 3 State achievable benchmark was 8.7 per 1,000 discharges.vi No racial/ethnic group or insurance group has met the achievable benchmark. Data are insufficient to determine time to benchmark.

Also, in the NHQR:

  • From 2008 to 2010, there were no statistically significant changes in the overall risk-adjusted rate of postoperative sepsis.
  • In 2009 and 2010, surgery patients ages 18-44 had lower risk-adjusted rates of postoperative sepsis than those ages 45-64 and those age 65 and over.
  • In 2008, surgery patients in hospitals with fewer than 100 beds had lower risk-adjusted rates of postoperative sepsis than those in hospitals with 500 or more beds. In 2009 and 2010, the rate was higher in the smallest hospitals (under 100 beds) than in the largest hospitals (500 or more beds).
  • As of 2010, the achievable benchmark had not been met for any age group or hospital bed size.  
Outcome: Catheter-Associated Urinary Tract Infections

The urinary tract is a common site of HAIs. Urinary catheter use and specific comorbid conditions can increase the risk of developing a UTI. Approximately 40% of all HAIs are attributed to catheter-associated UTIs (Niel-Weise & van den Broek, 2005).

Figure 4.3. Adult surgery patients with postoperative catheter-associated urinary tract infection, by race/ethnicity, 2009-2011

Select [D] Text Description below for details.

[D] Select for Text Description.

Source: Agency for Healthcare Research and Quality and Centers for Medicare & Medicaid Services, Medicare Patient Safety Monitoring System, 2009-2011.
Denominator: Selected discharges of hospitalized patients age 18 and over having major surgery and meeting specific criteria for each measure.
Note: For this measure, lower rates are better. White, Black, and Asian are non-Hispanic. Hispanic includes all races. Data for Asians and Hispanics in 2009 did not meet criteria for statistical reliability.

  • In 2010, the percentage of adult surgery patients with catheter-associated UTIs was higher for Hispanics than for Whites. There were no statistically significant differences among racial/ethnic groups in 2009 or 2011 (Figure 4.3).

Also, in the NHQR:

  • From 2009 to 2011, there were no statistically significant changes in the overall rate of postoperative catheter-associated UTIs.
  • In all years, the percentage of adult surgery patients with catheter-associated UTIs was higher for those ages 65-74, 75-84, and 85 and over than for adult surgery patients under age 65.
  • In all years, the percentage of adult surgery patients with catheter-associated UTIs was higher for patients with renal disease than for patients without renal disease.
Outcome: Central Line-Associated Bloodstream Infections

Patients who require a central venous catheter are severely ill. The use of these catheters increases the risk of serious complications such as bloodstream infections. CDC collects data about CLABSIs and presents the rates of infection by the teaching status of the hospital for some types of units. Because major teaching hospitals generally have higher proportions of disadvantaged populations than do nonteaching hospitals, results of this comparison are shown in the NHDR.

Figure 4.4. Rate per 1,000 central-line days of central line-associated bloodstream infections in major teaching hospitals and non-major teaching hospitals, 2006-2011

Select [D] Text Description below for details.

[D] Select for Text Description.

Source: Centers for Disease Control and Prevention, National Healthcare Safety Network annual reports, 2008-2011.
Denominator: Number of central-line days.
Note: For this measure, lower rates are better. Major teaching hospitals are hospitals that are an important part of a teaching program at a medical school.

  • CLABSI rates in medical intensive care units (ICUs) located in major teaching hospitals were 2.6 per 1,000 central-line days during the baseline period of 2006-2008 and 1.2 per 1,000 central-line days in 2011. In comparison, CLABSI rates in medical ICUs located in other types of hospitals were 1.9 per 1,000 central-line days in 2006-2008 and 1.1 per 1,000 central-line days in 2011 (Figure 4.4).
  • CLABSI rates in combined medical/surgical ICUs located in major teaching hospitals were 2.1 per 1,000 central-line days in 2006-2008 and 1.4 per 1,000 central-line days in 2011.

Also, in the NHQR:

  • From the referent period (2006-2008) to 2011, CLABSIs reported to the NHSN decreased by roughly 40%.
  • There was a significant decrease in CLABSIs among facilities that reported in both 2010 and 2011 and among facilities that reported in both 2009 and 2010. There was no statistically significant difference reported from January-June 2009 to July-December 2009 among facilities that reported for both time periods.

Other Adverse Events

Outcome: Mechanical Adverse Events Associated With Central Venous Catheters

Some patients need central venous catheters inserted into major veins in the neck, chest, or groin so that health care providers can administer medication or fluids, obtain blood for tests, or take cardiovascular measurements. Patients who require a central venous catheter tend to be severely ill. The placement and use of these catheters can result in mechanical adverse events, including bleeding; hematoma; perforation; pneumothorax; air embolism; and misplacement, occlusion, shearing, or knotting of the catheter.

Figure 4.5. Composite: Mechanical adverse events associated with central venous catheter placement among adults, by race and sex, 2009-2011

Select [D] Text Description below for details.

 

Select [D] Text Description below for details.

[D] Select for Text Description.

Source: Agency for Healthcare Research and Quality and Centers for Medicare & Medicaid Services, Medicare Patient Safety Monitoring System, 2009-2011.
Denominator: Selected discharges of hospitalized patients age 18 and over with central venous catheter placement.
Note: For this measure, lower rates are better. White and Black are non-Hispanic. Mechanical adverse events include allergic reaction to the catheter, tamponade, perforation, pneumothorax, hematoma, shearing off of the catheter, air embolism, misplaced catheter, thrombosis or embolism, knotting of the pulmonary artery catheter, and certain other events.

  • In all years from 2009 to 2011, there were no statistically significant differences by sex in the percentage of mechanical adverse events associated with central venous catheter placement (Figure 4.5).
  • Only in 2009 did Blacks have a higher percentage of mechanical adverse events associated with central venous catheter placement than Whites.

Also, in the NHQR:

  • From 2009 to 2011, there were no statistically significant changes in the overall percentage of mechanical adverse events associated with central venous catheter placement.
  • In all years, there were no statistically significant differences by age in the percentage of mechanical adverse events associated with central venous catheter placement.
Outcome: Obstetric Trauma

Childbirth and reproductive care are the most common reasons for women of childbearing age to use health care services. With an average of 10,957 babies born each day in the United States (Martin, et al., 2012), childbirth is the most common reason for hospital admission among women.

Obstetric trauma involving a severe tear to the vagina or surrounding perineal tissues during delivery is a frequent complication of childbirth. Higher risks of severe (i.e., 3rd or 4th degree) perineal laceration may be related to the degree of fetal-maternal size disproportion. Adolescents, who often have smaller body sizes, may be more likely to experience obstetric trauma than older women. In addition, although any delivery can result in trauma, existing evidence shows that severe perineal trauma can be reduced by restricting the use of episiotomies and forceps (Kudish, et al., 2008).

Previous reports used AHRQ Quality Indicators version 3.1 to generate obstetric trauma rates. As of the 2011 NHQR, the reports use a modified version 4.1 of the software. While the effects of this version change are extremely small, these estimates should not be compared with estimates found in previous reports.

Figure 4.6. Obstetric trauma with 3rd or 4th degree laceration per 1,000 vaginal deliveries without instrument assistance, by race/ethnicity and area income, 2004-2010

Select [D] Text Description below for details.

 

Select [D] Text Description below for details.

[D] Select for Text Description.

Key: API = Asian or Pacific Islander; Q1 represents the lowest income quartile and Q4 represents the highest income quartile based on the median income of a patient's ZIP Code of residence.
Source: Agency for Healthcare Research and Quality (AHRQ), Healthcare Cost and Utilization Project, Nationwide Inpatient Sample and AHRQ Quality Indicators, modified version 4.1, 2004-2010.
Denominator: All patients hospitalized for vaginal delivery without indication of instrument assistance.
Note: For this measure, lower rates are better. Rates are adjusted by age. White, Black, and API are non-Hispanic; Hispanic includes all races.

  • From 2004 to 2010, rates of obstetric trauma with 3rd or 4th degree laceration decreased for all racial/ethnic and area income groups (Figure 4.6).
  • In all years from 2004 to 2010, API mothers had higher rates of obstetric trauma than White mothers. In the same period, Black and Hispanic mothers had lower rates of obstetric trauma than White mothers.
  • In all years, residents in the upper three area income quartiles had higher rates than residents in the lowest area income quartile.
  • The 2008 top 4 State achievable benchmark was 17.8 per 1,000 deliveries.vii Hispanic and Black mothers and residents in the lowest area income quartile already have attained the benchmark. At the current annual rate of decrease, it could take up to 5 years for White mothers and up to 16 years for API mothers to reach the benchmark. All income groups could achieve the benchmark within about 7 years.

Also, in the NHQR:

  • From 2004 to 2010, the overall rate of obstetric trauma with 3rd or 4th degree laceration decreased from 30 to 22.8 per 1,000 vaginal deliveries without instrument assistance. The rates for all age groups and insurance types decreased.
  • In most years, mothers ages 18-24 had a lower rate of obstetric trauma than mothers ages 10-14, 15-17, and 25-34. Similarly, mothers with private insurance had higher rates of obstetric trauma than all other insurance types in almost every year.
  • At the current annual rate of decrease, the achievable benchmark could be attained within about 1 year by several age groups and insurance types. However, it could take up to 10 years for mothers under age 18 and ages 25-34, as well as for those with private insurance. Mothers with Medicare or Medicaid already have achieved the benchmark.

Nursing Home Setting

Outcome: Pressure Ulcers, Use of Restraints, and Urinary Tract Infections

People may seek nursing home care for short periods of time after hospitalization while they are recuperating to be able to return to their homes, or they may enter a nursing home permanently because they can no longer care for themselves at home. For both types of nursing home residents, optimal care seeks to maximize quality of life and minimize unintended complications.

Since 2002, CMS has collected data using the Minimum Data Set (MDS). The MDS provides data on nursing home residents at specified intervals during their stay that describe the resident's physical and clinical conditions. In 2010, nursing homes began reporting data using an updated instrument (MDS 3.0). We present the 2011 results for new quality measures developed for this version of the MDS that look at pressure ulcers, use of restraints, and UTIs.

A pressure ulcer, or pressure sore, is an area of soft tissue injury caused by sitting or lying in one position for an extended time and can be very painful and lead to infections. Nursing homes can do several things that may help to prevent or treat pressure sores, such as frequently changing the resident's position, providing proper nutrition, and using soft padding to reduce pressure on the skin.
Residents who are restrained daily can become weak, lose their ability to go to the bathroom by themselves, and develop pressure sores or other medical conditions. Restraints should only be used when medically necessary, and even then only under careful supervision.

Most UTIs can be prevented by keeping the genital area clean, emptying the bladder regularly, and drinking enough fluid. Finding the cause and getting early treatment of a UTI can prevent the infection from spreading and becoming more serious or causing complications.

Figure 4.7. Nursing home residents experiencing various adverse events, by race/ethnicity, 2011

Select [D] Text Description below for details.

[D] Select for Text Description.

Key: AI/AN = American Indian or Alaska Native; NHOPI = Native Hawaiian or Other Pacific Islander.
Source: Centers for Medicare & Medicaid Services, Minimum Data Set 3.0, 2011.
Denominator: For pressure ulcers, the denominator was short-stay residents, who are defined as residents whose cumulative stay was less than or equal to 100 days. For restraints and urinary tract infections, the denominator was long-stay residents, who are defined as residents whose cumulative stay was greater than 100 days.
Note: For these measures, lower rates are better. Measures were calculated as follows: Pressure ulcers: Percentage of short-stay residents for whom a look-back scan indicates one or more new or worsening stage II-IV pressure ulcers. Restraints: Percentage of long-stay residents who are physically restrained on a daily basis. UTI: Percentage of long-stay residents with a urinary tract infection within the 30 days prior to assessment.

  • In 2011, a higher percentage of Black short-stay residents had pressure ulcers compared with White residents. The percentage was lower for Asian and Hispanic residents than for White residents (Figure 4.7).
  • The percentage of long-stay residents with restraint use was higher for Asian, Native Hawaiian or Other Pacific Islander (NHOPI), Hispanic, and multiple-race residents than for White residents. The percentage was lower for Blacks than for Whites.
  • The percentage of long-stay residents with a UTI was higher for Whites than for other racial and ethnic groups except residents described as multiple race.

Also, in the NHQR:

  • In 2011, the percentage of short-stay residents with pressure ulcers was higher for males than for females.
  • There were no statistically significant differences by sex in the percentage of long-stay residents who were physically restrained.
  • The percentage of long-stay residents with UTIs was higher for females than for males.
  • The percentage of short-stay residents with pressure ulcers and the percentage of long-stay residents with UTIs increase with age. For both measures, the percentage of residents ages 65-74, 75-84, and 85 and over who met the criteria was greater than the percentage of residents ages 0-64.
  • Compared with long-stay residents under age 65, a lower percentage of long-stay residents ages 65-74 were physically restrained on a daily basis. There was no statistically significant difference in the percentage with restraint use between those under age 65 and those ages 75-84 or 85 and over.

Home Health Setting

Outcome: Improvement in Surgical Site Wound Healing

Normal wound healing after an operation is an important marker of good care. Patients whose wounds heal normally generally feel better and can get back to their daily activities sooner than those whose wounds do not heal normally. The home health team can assist with wound healing in several ways, including changing the wound dressing and teaching the patient or caregiver about wound healing (e.g., signs of wound healing, type of foods that promote wound healing, signs of infection, what to do about signs of infection or other concerns). One way to measure the quality of care that home health agencies give is to look at how well their patients' wounds heal after an operation.

Since 1999, CMS has required home health agencies to collect and report data using the Outcome and Assessment Information Set (OASIS). OASIS provides data on patients whose care is reimbursed by Medicare or Medicaid. Beginning in 2010, home health agencies used a revised version of the instrument called OASIS-C.

This outcome measure is derived from the OASIS-C dataset and describes the percentage of home health episodes where the status of the surgical incision site was better at the end of the home health care episode than at the start of the episode.

Figure 4.8. Home health patients with improvement in their surgical site wounds, by race and ethnicity, 2010-2011

Select [D] Text Description below for details.

[D] Select for Text Description.

Key: AI/AN = American Indian or Alaska Native; NHOPI = Native Hawaiian or Other Pacific Islander.
Source: Centers for Medicare & Medicaid Services, Outcome and Assessment Information Set, 2010-2011.
Denominator: Number of home health episodes during the measurement period in which the patient had a surgical wound and the episode ended with the patient discharged from home health care.
Note: For ethnicity, White, Black, and Other are non-Hispanic. Hispanic includes all races.

  • In 2010 and 2011, there were no statistically significant racial or ethnic differences in the percentage of home health patients with improvement in surgical site wound healing (Figure 4.8).
  • In 2011, the percentage with improved wound healing ranged from 86.2% for multiple-race home health patients to 89.1% for NHOPI home health patients.
  • In 2011, the percentage of non-Hispanic White home health patients with improvement in their surgical site wounds was 88.2%; the percentage of non-Hispanic Black home health patients was 87.2%; the percentage of non-Hispanic Other home health patients was 88.0%; and the percentage of Hispanic home health patients was 86.9%.

Also, in the NHQR:

  • The overall percentage of home health patients with improvement in their surgical site wound healing was 85.9% in 2010 and 87.9% in 2011.
  • In 2011, 84.7% of home health patients ages 0-64 had improvement in their surgical site wounds; for home health patients ages 65-74, 88.2% had improvement. The percentage of home health patients ages 75-84 with improvement was 89.4%, and the percentage of patients age 85 and over was 90.4%.

Outcome: Ability To Take Medications Orally

The ability to perform daily activities, such as taking medications correctly, is important to the health status and quality of life of people living in the community. Taking too much or too little can keep the medications from working properly and may cause unintended harm, including death.

The home health team can help teach patients ways to organize medications and to take them properly. Getting better at taking medications correctly means the home health team is doing a good job teaching patients how to take their medications and about the harm that can occur if they do not follow these instructions. Specific items that should be discussed include all the prescriptions and other medications the patient takes, allergic or other adverse reactions to medications experienced in the past, and actions to take if a medication is not working.

This measure shows how often the home health team helped patients get better at taking their prescription and other medications correctly (including prescription medications, over-the-counter medications, vitamins, and herbal supplements). Only medications the patient takes by mouth are considered.

Figure 4.9. Home health patients with improvement in their ability to take medications orally, by race and ethnicity, 2010-2011

Select [D] Text Description below for details.

[D] Select for Text Description.

Key: AI/AN = American Indian or Alaska Native; NHOPI = Native Hawaiian or Other Pacific Islander.
Source: Centers for Medicare & Medicaid Services, Outcome and Assessment Information Set, 2010-2011.
Denominator: Number of home health episodes of care in which a patient was unable to take oral medications independently at the start of the episode that ended during the measurement period.
Note: For ethnicity, White, Black, and Other are non-Hispanic. Hispanic includes all races.

  • The percentage of home health patients with improvement in their ability to take oral medications was significantly lower for Hispanics than for non-Hispanic Whites in 2010 and 2011 (Figure 4.9). In 2010, 37.4% of Hispanic home health patients showed improvement in their ability to take medications orally compared with 46.9% of non-Hispanic White home health patients. In 2011, 36.8% of Hispanic home health patients showed improvement compared with 48.4% of non-Hispanic White home health patients.
  • In both years, there were no statistically significant racial differences in the percentage of home health patients with improvement in their ability to take medications orally. In 2011, the percentages were: White, 48.4%; Black, 49.3%; Asian, 45.1%; NHOPI, 47.0%; American Indian or Alaska Native (AI/AN), 46.6%; and multiple race, 48.5%.

National data suggest that Hispanics have significantly lower rates of improvement in their ability to take medications orally compared with non-Hispanic Whites. The figure below examines this disparity by State.

Figure 4.10. Improvement in the ability of home health patients to take medications orally, by ethnicity and State, 2011

Select [D] Text Description below for details.

 

[D] Select for Text Description.

  • In Nevada, Texas, North Carolina, and Florida, the percentage of patients with improvement in their ability to take medications orally was significantly lower for Hispanics than for non-Hispanic Whites (Figure 4.10).
  • In Washington and South Carolina, the percentage of patients with improvement in their ability to take medications orally was significantly higher for Hispanics than for non-Hispanic Whites.

Also, in the NHQR:

  • The overall percentage of home health patients with improvement in their ability to take medications orally was 46.2% in 2010 and 47.3% in 2011.
  • In 2010 and 2011, the percentage of home health patients with improvement in their ability to take medications orally was significantly lower for those age 85 and over compared with those ages 0-64.

Ambulatory Care Setting

Outcome: Ambulatory Care Visits Due to Adverse Effects of Medical Care

Although patient safety initiatives are predominantly focused on inpatient hospital events, many adverse effects of medical care will be treated during visits to outpatient settings. Outpatient providers who see patients experiencing adverse effects of medical care include physician offices, urgent care centers, ambulatory surgery centers, and hospital outpatient departments. Patient safety events that are identified and treated in the ambulatory setting may also originate from that setting.

While most ambulatory care is less technologically complex than inpatient care, it is often more complex logistically, potentially involving failures in communication and coordination among a number of providers and locations. Some adverse effects, such as known side effects of appropriately prescribed medications, may be unavoidable, while others may be considered avoidable medical errors.

Figure 4.11. Ambulatory care visits due to adverse effects of medical care per 1,000 people, by race and geographic area, 2006-2009

Select [D] Text Description below for details.

 

Select [D] Text Description below for details.

[D] Select for Text Description.

Key: MSA = metropolitan statistical area.
Source: Centers for Disease Control and Prevention, National Center for Health Statistics, National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey, 2006-2009.
Denominator: U.S. Census Bureau estimated civilian noninstitutionalized population as of July 1 of each data year.
Note: For this measure, lower rates are better. Ambulatory care includes visits to office-based physicians, hospital outpatient departments, and hospital emergency departments.

  • For the 2-year rolling averages shown, there were no statistically significant racial differences in the rate of ambulatory care visits due to adverse effects of medical care (Figure 4.11).
  • Only in 2007-2008 was the rate of ambulatory care visits due to adverse effects of medical care different (higher) for residents of metropolitan areas compared with residents of nonmetropolitan areas (44.1 vs. 28.8 visits per 1,000 population, respectively).

Also, in the NHQR:

  • There was no statistically significant difference between 2006-2007 and 2008-2009 in the rate of ambulatory care visits due to adverse effects of medical care.
  • In all three time periods, the rates of ambulatory care visits due to adverse effects were higher for patients ages 18-44, 45-64, and 65 and over than for patients ages 0-17.
  • The rate of ambulatory care visits due to adverse effects of medical care was higher for females compared with males in all three time periods.

Outcome: Receipt of Potentially Inappropriate Prescription Medications

Some medications are potentially harmful for older patients but still are prescribed to them (Zhan, et al., 2001).viii Using inappropriate medications can be life threatening and may result in hospitalization, as well as increased costs of pharmaceutical services (Lau, et al., 2005). Measures of inappropriate medication use include the Beers criteria, which have been generally accepted by the medical community and by expert opinion, although there is still some disagreement (Fick, et al., 2012). This disagreement relates to the many factors that must be considered when identifying what constitutes inappropriate use by certain populations (Zhan, et al., 2001).

Figure 4.12. Adults age 65 and over who received potentially inappropriate prescription medications in the calendar year, by race/ethnicity and family income, 2002-2010

Select [D] Text Description below for details.

 

Select [D] Text Description below for details.

[D] Select for Text Description.

Source: Agency for Healthcare Research and Quality, Medical Expenditure Panel Survey, 2002-2010.
Denominator: Civilian noninstitutionalized population age 65 and over.
Note: For this measure, lower rates are better. Prescription medications received include all prescribed medications initially purchased or otherwise obtained, as well as any refills. White and Black are non-Hispanic. Hispanic includes all races.

  • The percentage of older adults who received potentially inappropriate medications decreased for all racial/ethnic and family income groups from 2002 to 2010 (Figure 4.12).
  • In all years, there were no statistically significant racial/ethnic differences in the percentage of adults age 65 and over who received potentially inappropriate medications.
  • In most years, there were no statistically significant income differences in the percentage of adults age 65 and over who received potentially inappropriate medications.

Also, in the NHQR:

  • From 2002 to 2010, the overall percentage of adults age 65 and over who received potentially inappropriate medications decreased. The percentages for all age groups (except those ages 65-69) and all health status groups also declined.
  • Except in 2002, there were no statistically significant differences between adults ages 65-69 and older age groups in the percentage who received potentially inappropriate medications.
  • In all years, the percentage of adults age 65 and over who received potentially inappropriate medications was higher for those with fair/poor perceived health than for those with excellent/very good/good perceived health.

Outcome: Hospital Readmissions

One aim of the National Quality Strategy (NQS) is to make care safer by reducing the harm caused in the delivery of care. One of the two measures that the NQS has endorsed to describe improved safety is an all-payer 30-day readmission rate. The baseline rate calculated for the all-payer 30-day readmission rate in 2010 was 14.4% based on 32.9 million admissions. The goal is to reduce this rate by 20% by the end of 2014. In 2011, the rate was 14.4% based on 32.7 million admissions (HHS, 2013).

In addition, for certain diseases (acute myocardial infarctions [i.e., heart attacks], heart failure, and pneumonia), CMS's Hospital Quality Alliance (2008-2010)ix tracked and published 30-day risk-standardized readmission rates among Medicare fee-for-service patients age 65 and over in hospitals on their Hospital Compare Web site. Rates of readmission may reflect hospital efforts to prevent complications, teach patients at discharge, and ensure that patients make a smooth transition to their home or another setting such as a nursing home.

Figure 4.13. Median hospital 30-day risk standardized readmission rate, by proportion of African American and Medicaid patients, 2008-2010

Select [D] Text Description below for details.

 

Select [D] Text Description below for details.

[D] Select for Text Description.

Key: AMI = acute myocardial infarction.
Source: Hospital Compare Chartbook, 2012.
Denominator: Expected number of readmissions for each disease type given the hospital's case mix.
Note: For this measure, lower rates are better. For a hospital's proportion of patients who are African American, low is defined as 0% for all three measures. High is defined as ≥22% for AMI, ≥23% for heart failure, and ≥22% for pneumonia. For the proportion of the hospital's patients who are insured by Medicaid, low is defined as ≤8% for AMI, ≤7% for heart failure, and ≤6% for pneumonia. High is defined as ≥30% for AMI, ≥29% for heart failure, and ≥29% for pneumonia.

  • There were no statistically significant differences between hospitals serving a high percentage of African Americans and hospitals serving a low percentage of African Americans in the median 30-day risk-standardized readmission rates for all three diseases (Figure 4.13).
  • The median 30-day risk-standardized readmission rate for acute myocardial infarction was 20.4% in hospitals that served a high percentage of African American patientscompared with 19.2% in hospitals that served a low percentage of African American patients. For heart failure, the rates were 25.8% and 24.4%, respectively. For pneumonia, the rates were 19.1% and 17.9%, respectively.
  • There were no statistically significant differences between hospitals serving a high percentage of Medicaid recipients and hospitals serving a low percentage of Medicaid recipients in the median 30-day risk-standardized readmission rates for all three diseases.
  • The median 30-day risk-standardized readmission rate for acute myocardial infarction was 20.2% in hospitals that served a high percentage of Medicaid recipients compared with 19.5% in hospitals that served a low percentage of Medicaid recipients. For heart failure, the rates were 25.2% and 24.5%, respectively. For pneumonia, the rates were 18.7% and 17.9%, respectively.

Also, in the NHQR:

  • The median 30-day risk-standardized readmission rates for hospitals remained stable from 2006 to 2010 for all three diseases.

Patient Safety Infrastructure

The patient safety infrastructure also plays a role in making care better. This infrastructure includes scientific research, lessons from data-driven investigations, cultural shifts, and other actions taken to make care safer. This section highlights two hallmark activities in this regard:

  • An innovative State reporting system that allows learning from near misses and unsafe conditions.
  • An assessment of the patient safety culture in U.S. hospitals.

Patient Safety Event Reporting in Pennsylvania

In June 2004, Pennsylvania began implementing a statewide mandatory reporting system for patient safety events. All hospitals, ambulatory surgical facilities, birthing centers, and abortion facilities licensed by the State report through the Pennsylvania Patient Safety Reporting System. Pennsylvania was the first State to require the reporting of both patient safety events that cause harm and those events, such as "near-misses" or "unsafe conditions," that do not result in patient harm. The mandatory reporting laws and longstanding patient safety culture in the State provide a unique opportunity to view disparities in patient safety events among subgroups.

Figure 4.14 shows the variation in patient age for all hospital and ambulatory surgical facility event reports for the top three reported patient safety event categories in 2012 overall. 

Figure 4.14. Pennsylvania patient safety reports, by event type and patient age, 2012

Select [D] Text Description below for details.

[D] Select for Text Description.

Source: Pennsylvania Patient Safety Reporting System, 2012.
Note: Only hospital and ambulatory surgical facility data are presented.

  • In 2012, for errors related to procedures, treatment, or tests, most reported events (50.5%) occurred among adults ages 18-64. This was followed by adults age 65 and over (40%). In contrast, infants under the age of 1 (including newborns) and children ages 1-17 experienced 4.7% and 4.8%, respectively, of these errors (Figure 4.14).
  • For medication errors, the largest number of reported events (43.9 %) also occurred among adults ages 18-64. This was closely followed by adults age 65 and over (43.1%). In contrast, infants under the age of 1 (including newborns) experienced 4.5% percent of these errors, and children ages 1-17 experienced 8.5 % of all medication errors.
  • Adults age 18 and over accounted for nearly all reported falls. The majority (52.0%) occurred among those age 65 and over, while adults ages 18-64 also made up a large percentage of falls (45.0%).

Figure 4.15. Pennsylvania patient safety reports by harm and age, 2012

Select [D] Text Description below for details.

[D] Select for Text Description.

Source: Pennsylvania Patient Safety Reporting System, 2012.
Note: Only hospital and ambulatory surgical facility data are presented.

  • In 2012, most (96.5%) of the more than 235,000 patient safety events reported by Pennsylvania hospitals and ambulatory surgical facilities did not involve patient harm (data not shown). The Pennsylvania Patient Safety Authority's collection and analysis of information reported about near misses and other no-harm events is an essential component of learning how to minimize patient safety events that cause harm.
  • The percentage of events that were near misses was higher for children and teens ages 1-17 than for those involving other age groups (Figure 4.15). A near miss is an event that did not reach the individual because of either chance alone or active recovery efforts by caregivers.
  • Of the events reported in 2012 that involved infants under the age of 1, 1.6% caused patient harm or death. Patient harm or death was somewhat more likely to occur among reported events involving older children, adults ages 18-64, and adults age 65 and over. Patient harm or death was the result for 3.5% of reported events involving children ages 1-17, 3.7% for adults ages 18-64, and 3.3% for those age 65 and over.

Outcome: Patient Safety Culture

High-reliability organizations—those that perform high-risk work but achieve low rates of adverse events—establish "cultures of safety." A culture of safety is characterized by shared dedication to making work safe, nonpunitive reporting and communication about error, collaboration and teamwork across disciplines, and adequate resources to prevent adverse events. AHRQ developed the Hospital Survey on Patient Safety Culture to help hospitals assess the culture of safety in their facilities. The 2013 NHDR presents data from the Hospital Survey on Patient Safety Culture: 2012 User Comparative Database Report.

This report is based on survey responses collected from 567,703 hospital staff in 1,128 hospitals that represent approximately 18% of the country's hospitals. The average hospital response rate was 53%, with an average of 503 completed surveys per participating hospital. Hospitals contributing data to the comparative database mirror the population of U.S. hospitals as a whole, but participation is entirely voluntary.

Most hospitals administered Web surveys (66%). Web surveys resulted in slightly lower response rates (51%) than response rates from paper surveys (61%) but were about the same as mixed-mode administered surveys (49%). Most hospitals administered the survey to all staff or to a sample of all staff from all hospital departments. Nurses accounted for more than one-third of respondents. More than three-quarters of respondents had direct interaction with patients.

The survey assesses 12 patient safety culture composites, as an average percent positive response. Percent positive refers to the percentage of responses that agree or strongly agree with a positively worded item (e.g., "People support one another in this unit") and the percentage that disagree or disagree strongly with a negatively worded item (e.g., "We have patient safety problems in this unit").

Figure 4.16. Average percent positive response for teamwork across units and handoffs and transitions, by hospital teaching status, 2012

Select [D] Text Description below for details.

[D] Select for Text Description.

Source: Agency for Healthcare Research and Quality, Hospital Survey on Patient Safety Culture, 2012.

  • Nearly two-thirds (66%) of the database hospitals were nonteaching, which is slightly lower than the proportion of 2010 American Hospital Association registered hospitals (76%) that were nonteaching. Nonteaching hospitals scored higher than teaching hospitals across all 12 patient safety culture composites (data not shown).
  • In particular, nonteaching hospitals on average scored higher by 5 percentage points on Teamwork Across Units (60% positive compared with 55%) and Handoffs and Transitions (47% positive compared with 42%) (Figure 4.16).

 

References

 2013 annual progress report to Congress: National Strategy for Quality Improvement in Health Care. Washington, DC: U.S. Department of Health and Human Services; July 2013. Available at: https://www.ahrq.gov/workingforquality/nqs/nqs2013annlrpt.pdf (PDF File, 1.3 MB). Accessed July 24, 2013.

 2012 annual progress report to Congress. National Strategy for Quality Improvement in Health Care. Washington, DC: U.S. Department of Health and Human Services; April 2012, corrected August 2012. Available at: https://www.ahrq.gov/workingforquality/nqs/nqs2012annlrpt.pdf  (PDF File, 408 KB). Accessed May 13, 2013.

 Healthcare-associated infections (HAIs) data and statistics. Atlanta, GA: Centers for Disease Control and Prevention. December 13, 2010. Available at: http://www.cdc.gov/HAI/surveillance/index.html. Accessed April 15, 2014.

 Fick D, Semla T, Beizer J, et al. American Geriatrics Society updated Beers criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc 2012; 60(4):616-31.

 Gastmeier P. Nosocomial infection surveillance and control policies. Curr Opin Infect Dis 2004 Aug;17(4):295-301.

 Jha AK, Chan DC, Ridgway AB, et al. Improving safety and eliminating redundant tests: cutting costs in U.S. hospitals. Health Aff (Milwood) 2009 Sep-Oct;28(5):1475-84.

 Kohn L, Corrigan J, Donaldson M, eds. To err is human: building a safer health system. Institute of Medicine, Committee on Quality of Health Care in America. Washington, DC: National Academy Press; 2000.

 Kudish B, Sokol RJ, Kruger M. Trends in major modifiable risk factors for severe perineal trauma, 1996-2006. Int J Gynaecol Obstet 2008 Aug; 102(2):165-70.

 Landrigan CP, Parry GJ, Bones CB, et al. Temporal trends in rates of patient harm resulting from medical care. N Engl J Med 2010 Nov 25;363(22):2124-34. Erratum in: N Engl J Med. 2010 Dec 23;363(26):2573.

 Lau DT, Kasper JD, Potter DE, et al. Hospitalization and death associated with potentially inappropriate medication prescriptions among elderly nursing home residents. Arch Intern Med 2005 Jan 10;165(1):68-74.

 Martin JA, Hamilton BE, Ventura SJ, et al. Births: final data for 2010. Natl Vital Stat Rep 2012 Aug 28;61(1).

 Mayr FB, Yende S, Linde-Zwirble WT, et al. Infection rate and acute organ dysfunction risk as explanations for racial differences in severe sepsis. JAMA 2010 Jun 23;303(24):2495-2503.

 Metersky ML, Hunt DR, Kliman R, et al. Racial disparities in the frequency of patient safety events: results from the National Medicare Patient Safety Monitoring System. Med Care 2011 May;49(5):504-10.

 Moore LJ, Moore FA, Todd SR, et al. Sepsis in general surgery: the 2005-2007 National Surgical Quality Improvement Program perspective. Arch Surg 2010 Jul;145(7):695-700.

 National Academy for State Health Policy. Patient Safety Toolbox. Available at: http://www.nashp.org/pst-welcome. Accessed May 13, 2013.

 Niel-Weise BS, van den Broek PJ. Antibiotic policies for short-term catheter bladder drainage in adults. Cochrane Database Syst Rev 2005;3.

 Raetzman SO, Fullerton C, Frankel S. Review to support AHRQ report on effective strategies for improving patient safety.(Prepared by Truven Health Analytics for AHRQ under the Network of Patient Safety Databases contract.) Rockville, MD: Agency for Healthcare Research and Quality; November 26, 2012.

 Shreve J, Van Den Bos J, Gray T, et al. The economic measurement of medical errors. Schaumberg, IL: Society of Actuaries/Milliman; 2010.

 Zhan C, Sangl J, Bierman AS, et al. Potentially inappropriate medication use in the community-dwelling elderly: findings from the 1996 Medical Expenditure Panel Survey. JAMA 2001 Dec 12;286(22):2823-9.

 Zimlichman E, Henderson D, Tamir O, et al. A meta-analysis of costs and financial impact on the U.S. health care system. JAMA InternMed 2013 Dec 9-23;173(22):2039-46.


i Some argue that this estimate may be too high, while others argue the estimate is too low because diagnosis-related errors are not counted here. See diagnosis-related errors section below.
ii The Jha, et al., estimate for 2004 at $16.622 billion was inflated by the Producer Price Index for medical and surgical hospitals through 2012, plus an assumed 2.2% increase for 2013.
iii Available at https://www.ahrq.gov/workingforquality/reports.htm.
iv MPSMS data were based solely on patients from four patient groups whose charts were requested by CMS for hospitals as part of the Hospital Inpatient Quality Reporting Program. These groups included all-payer patients >17 years old with a principal discharge diagnosis of: (1) pneumonia, (2) acute myocardial infarction, (3) heart failure, or (4) major surgery (as described in the Surgical Care Improvement Project ([SCIP]).
v All Other HACs includes: inadvertent femoral artery puncture for catheter angiographic procedures, adverse event associated with hip joint replacement, adverse event associated with knee joint replacement, contrast nephropathy associated with catheter angiography, hospital-acquired methicillin-resistant Staphylococcus aureus (MRSA), hospital-acquired vancomycin-resistant Enterococcus (VRE), hospital-acquired antibiotic-associated Clostridium difficile, mechanical complications associated with central venous catheters, postoperative cardiac events for cardiac and noncardiac surgeries, postoperative pneumonia, iatrogenic pneumothorax (HCUP Patient Safety Indicator [PSI] 6), postoperative hemorrhage or hematoma (PSI 9), postoperative respiratory failure (PSI 11), and accidental puncture or laceration (PSI 15).
vi The top 3 States that contributed to the achievable benchmark are Nebraska, New Hampshire, and Rhode Island.
vii The top 4 States that contributed to the achievable benchmark are South Dakota, Utah, West Virginia, and Wyoming.
viii Medications that should always be avoided for older patients include barbiturates, flurazepam, meprobamate, chlorpropamide, meperidine, pentazocine, trimethobenzamide, belladonna alkaloids, dicyclomine, hyoscyamine, and propantheline. Medications that are rarely appropriate for older patients or sometimes indicated for older patients but often misused include carisoprodol, chlorzoxazone, cyclobenzaprine, metaxalone, methocarbamol, amitriptyline, chlordiazepoxide, diazepam, doxepin, indomethacin, dipyridamole, ticlopidine, methyldopa, reserpine, disopyramide, oxybutynin, chlorpheniramine, cyproheptadine, diphenhydramine, hydroxyzine, promethazine, and propoxyphene.
ix The Hospital Quality Alliance no longer exists but did collect and report data for 2008-2010 used on the Hospital Compare Web site. Future data on readmissions will come from the Hospital Inpatient Quality Reporting Program.


Return to Contents

Page last reviewed May 2014
Page originally created May 2014
Internet Citation: Chapter 4. Patient Safety. Content last reviewed May 2014. Agency for Healthcare Research and Quality, Rockville, MD. https://archive.ahrq.gov/research/findings/nhqrdr/nhdr13/chap4.html

 

The information on this page is archived and provided for reference purposes only.

 

AHRQ Advancing Excellence in Health Care